Title: Hybrid fuzzy k-nearest neighbour technique for assessing the level of interior acoustics in vehicle cabin
Authors: Ahmad Kadri Junoh
Addresses: Industrial Mathematics Research Group, Institute of Engineering Mathematics, Universiti Malaysia Perlis, 02600, Arau, Perlis, Malaysia
Abstract: Interior acoustical comfort in the cabin is an element that may attract customers before they purchase a vehicle. In order to assess certain level of acoustical comfort, many studies have been performed in order to identify and classify acoustics level. Since a passenger may find it difficult to express the level of experienced noise numerically, a model of hybrid fuzzy k-nearest neighbour is used to classify the qualitative noise level into a few categories of annoyance levels. Later, the classifier of the linear discriminant analysis was used in order to validate the results produced. The finding of the studies on the developed model is able to assess the level of acoustics by classifying them into numerical scales that are meaningful for assessment.
Keywords: sound quality; acoustics level; fuzzy KNN; k-nearest neighbour; FKNN; linear discriminant analysis; LDA; interior acoustics; vehicle cabins; acoustical comfort.
International Journal of Vehicle Noise and Vibration, 2015 Vol.11 No.3/4, pp.225 - 237
Received: 20 Feb 2015
Accepted: 25 Jul 2015
Published online: 05 Mar 2016 *